LLA-FLOW: A Lightweight Local Aggregation on Cost Volume for Optical Flow Estimation
This work addresses deployment challenges in optical flow estimation for computer vision applications, but it is incremental as it builds on existing local constraint ideas.
The paper tackled the problem of high memory and time costs in optical flow estimation by proposing lightweight local aggregation modules (LSA and SLSA) for cost volumes, achieving competitive performance on the Sintel benchmark with reduced computational requirements.
Lack of texture often causes ambiguity in matching, and handling this issue is an important challenge in optical flow estimation. Some methods insert stacked transformer modules that allow the network to use global information of cost volume for estimation. But the global information aggregation often incurs serious memory and time costs during training and inference, which hinders model deployment. We draw inspiration from the traditional local region constraint and design the local similarity aggregation (LSA) and the shifted local similarity aggregation (SLSA). The aggregation for cost volume is implemented with lightweight modules that act on the feature maps. Experiments on the final pass of Sintel show the lower cost required for our approach while maintaining competitive performance.